Different methods for detecting chemical and biological analytes have been used. Such technology has been used, for example, in process control, environmental monitoring, medical diagnostics and security.
Mass spectroscopy is one approach to detect such analytes. The process begins with an ionized sample. The ionized sample is shot through a vacuum that is subjected to an electromagnetic field. The electromagnetic field changes the path of lighter ions more than heavier ions. A series of detectors or a photographic plate are then used to sort the ions depending on their mass. The output of this process, which is the signal from the detectors or the photographic plate, can be used to determine the composition of the analytes in the sample.
A disadvantage of mass spectroscopy instruments is that they are generally high-cost instruments. Additionally, they are difficult to ruggedize, and are not useful for applications that require a sensor head to be remote from signal-processing electronics.
A more recent approach is to use Micro Electro Mechanical Systems (MEMS)-based microstructures, and more specifically micro-cantilevers. These are extremely sensitive systems, and several demonstrations of mass sensors that have detection limits as low 10−21 g, approximately the mass of a single protein molecule, have been performed. While these experiments have been performed in idealised environments, practical cantilever-based systems have been demonstrated for the detection of a wide range of single analytes.
With these sensors, a portion of the micro-cantilever is coated with an analyte selective coating to which the analyte is adsorbed. There are two common modes of operation of micro-cantilever sensors, namely static and dynamic.
In the static mode, a stress differential is induced across the cantilever due to preferential adsorption of an analyte onto the analyte selective coating causing the cantilever to bend. The extent of the bending is in direct relation to the amount of analyte adsorbed. The stress differential can be induced by the analyte causing swelling of an overlayer, or by changes in the Gibbs free energy of the surface.
In the dynamic mode, the adsorbed analyte changes the mass of the cantilever and hence its mechanical resonance frequency. The rate and size of the change in resonance frequency is then measured to estimate the analyte concentration. Active sensing using these structures is achieved by resonant excitation.
In general, long, compliant cantilevers are required for sensitive static sensors, while high sensitivity for dynamic sensors dictate that short, stiff beams with high Q-factor mechanical resonances are needed. The most sensitive MEMS-based sensors to date have been based on measurements of resonant frequency.
Readout technologies used with micro-cantilever sensors are primarily based on optical techniques developed for atomic force microscopy (AFM) analysis. Here, light is reflected from the cantilever tip to a distant quadrant detector, which process is referred to as optical leveraging. Electrical sensing and optical sensing techniques are also used. Electrical sensing includes piezoresistive, piezoelectric, capacitive, Lorentz force/emf sensing and tunneling current techniques. Optical sensing techniques include optical sensing based on optical interference, the optical interference being either in an interferometer or in the use of diffraction from an optical grating formed by a line of cantilevers. This latter configuration using an optical grating formed by a line of cantilevers is often described as an array in literature, but is still effectively a sensor for a single analyte.
Another approach to analyte detection is where large, compact, integrated arrays of individual sensors are used, particularly for multi-analyte, multi-analysis applications. These are particularly useful when an unknown substance is to be identified or if there is a number of chemical species to be tested for simultaneously. Examples of such requirements can be found in the screening of food for pesticide residues where there are many different potential contaminants, detection of different antibodies in a single blood sample, or the presence of any of the many possible illicit drugs or explosives in luggage. Additionally, an array of sensors can also give significantly improved statistics of detection (including fewer false-positives and false-negatives) by averaging the response over a large number of sensors, and allows the use of multivariate statistical chemometric techniques, as are typically applied in spectroscopic analysis.
There are several disadvantages with the sensors of today. There is, for example, a lack of compact, robust and cost-effective read-out technology that combines high sensitivity with high dynamic range. Sensors that are good at detecting small amounts of analyte typically have poor dynamic range which is especially noticeable when the levels of analyte are large. A problem with AFM-based cantilever systems is that they are very large as they incorporate bulky free space optics requiring a sensor for each cantilever output. A problem with electrical cantilever systems is that they require extensive power on-chip electronics
As is known in the art, an Atomic Force Microscope (AFM) consists of a cantilever with a pointed tip or probe at its end that is used to scan a sample surface. The cantilever is typically made of silicon or silicon nitride with a tip radius of curvature in the order of nanometers using micro-electromechanical fabrication techniques. When the tip is brought into proximity of the sample surface, forces between the tip and the sample lead to a deflection of the cantilever according to Hooke's law.
Interatomic forces between the probe tip and the sample surface cause the cantilever to deflect as the sample's surface topography (or other properties) change as the tip is scanned across the sample. A laser light reflected from the back of the cantilever measures the deflection of the cantilever. This information is fed back to a computer, which generates a map of topography and/or other properties of interest.
Various measurements can be made including measuring either the deflection of the cantilever (static mode) or a vibration frequency of the cantilever (dynamic mode). In some applications, the tip is coated with a thin film of ferromagnetic material that reacts to magnetic areas on the sample surface. Some applications include:                Measuring 3-dimensional topography of an integrated circuit device        Roughness measurements for chemical mechanical polishing        Analysis of microscopic phase distribution in polymers        Mechanical and physical property measurements for thin films        Imaging magnetic domains on digital storage media        Imaging of submicron phases in metals        Defect imaging in IC failure analysis        Microscopic imaging of fragile biological samples        Metrology for compact disk stampers        
A problem with current AFMs is that the sensitivity is limited by shot noise in the optical detection system. Although Brownian motion of the cantilever is a contributor to the noise, in practice it is not a factor as the shot noise is substantially greater than the noise induced by Brownian motion. While noise induced by Brownian motion may be reduced by cooling the cantilever, this is not practical for current AFMs as it may interfere with the alignment of the optical system. A further problem is that, the process of measuring an entire surface of a sample is time consuming, as the probe tip must make many passes over the sample in order to build up an image.
Yet a further problem with current AFMs is that the probe often needs to be replaced, and each time the probe is replaced the optical detection system needs to be re-calibrated, which is a time consuming process.
There is therefore a need for an improved system and method of using cantilevers to perform sensitive measurements.